Large population surveys often show gay men and lesbian women in similar ranges, with small gaps driven by age, location, and how the question is asked.
People ask this because a clean “more X than Y” answer feels tidy. Real data is messier. There’s no census roster of sexual orientation, so researchers rely on surveys, and each survey uses its own wording and sampling.
Are There More Gay Men Or Lesbian Women? What Data Sets Say
Across many national surveys, the share of men who identify as gay and the share of women who identify as lesbian often land close to each other. In some places, men come out a bit higher; in others, women do. In early U.S. National Health Interview Survey reporting on sexual identity, researchers noted no clear sex gap in the share identifying as gay/lesbian, while women were more likely to identify as bisexual.
Two guardrails keep the reading honest:
- Most official surveys are not built to “rank” gay men against lesbian women. They aim to measure sexual orientation as one item in a larger statistical program.
- Small groups mean wider uncertainty. A few tenths of a percent can flip the ordering, even with strong sampling.
Why This Question Is Harder Than It Sounds
Before you compare counts, you have to pin down what “gay” and “lesbian” means in the data. Surveys can measure three different things:
- Identity (labels people choose for themselves)
- Attraction (who someone feels drawn to)
- Behavior (who someone has had sex with)
Those do not line up one-to-one for every person. That’s normal. It also means two studies can disagree while both are grounded in real answers.
Survey Wording Can Shift The Count
Some questionnaires give separate labels (gay, lesbian, bisexual). Others use one combined option like “gay or lesbian,” then report results by sex. Some include “something else,” “don’t know,” or “prefer not to answer,” which changes how analysts group totals.
What Large Surveys Commonly Find By Sex
When surveys split results by sex, one pattern shows up often: men who are not straight are more likely to select “gay,” while women who are not straight are more likely to select “bisexual.” That does not make “lesbian” a tiny category. It means the label mix differs by sex, and the bisexual share among women can be larger in some age bands.
This is why “LGB overall” can tilt male in one report while “gay vs lesbian” stays close. The bisexual share can do most of the moving.
How To Read A Headline Stat Without Getting Misled
Use this checklist before you accept a claim that one group is “bigger.”
- Population and ages: adults 18+, 18–64, or a youth-only sample will not match.
- Measure used: identity, attraction, or behavior.
- Response options: what labels were offered, and what sat in “other/unknown.”
- Missing data: did the analysis drop “prefer not” responses or keep them in the denominator?
Table: How Major Data Sources Measure Gay And Lesbian Identity
The sources below publish methods and regular updates. The last column flags what each source can answer cleanly about gay men and lesbian women.
| Data Source | How It Measures Sexual Orientation | What You Can Compare |
|---|---|---|
| U.S. NHIS (CDC/NCHS) | Self-identified sexual identity within a national health survey | Gay men vs lesbian women shares by sex, with published methods and cautions |
| U.S. NHANES (CDC/NCHS) | Includes sexual identity and other measures in a health and exam survey | Identity patterns tied to health indicators, not a simple “who is larger” tally |
| U.S. NSFG (CDC/NCHS) | Measures identity, attraction, and behavior in a family growth survey | How identity, attraction, and behavior differ by sex |
| UK Annual Population Survey (ONS) | “Gay or lesbian” reported within a large household survey | LGB totals and “gay or lesbian” by sex, with official breakdowns |
| Canada CCHS (Statistics Canada) | Self-reported sexual orientation within a health survey | Gay and lesbian identity by sex, plus age splits |
| Large probability polls (Gallup) | Self-identification in national polling with trend tracking | Trends and label mix across time, with methods summaries |
| National censuses (varies) | Often measures same-sex couples, not orientation directly | Couple counts by sex pairing, not total gay/lesbian population |
| Academic meta-studies | Combines multiple surveys with aligned definitions | Cross-country comparisons when methods match |
What U.S. Health Surveys Suggest
The U.S. National Center for Health Statistics added a sexual identity measure to the National Health Interview Survey, which made it possible to report national shares by identity labels. In the first year of that measure, the report noted that the share of men and women identifying as gay/lesbian did not show a clear difference, while women were more likely to identify as bisexual. That single detail is a useful lens for the whole topic: the gay-versus-lesbian comparison can be close, while the bisexual share shifts the broader totals by sex.
If you want the exact question wording and category definitions, start with the NHIS sexual orientation FAQs.
One nuance worth noticing: some U.S. instruments use “gay” as the label for men and “lesbian or gay” for women. That choice is documented in published technical notes and it can affect how people map their identity onto the menu.
What UK Population Statistics Show
The UK’s Office for National Statistics publishes a yearly bulletin on sexual orientation in the UK based on the Annual Population Survey. Their reporting uses one option labeled “gay or lesbian” and breaks results down by sex. In the 2023 bulletin, men were more likely than women to identify as lesbian, gay or bisexual overall, and the bulletin also shows a clear age gradient, with younger groups reporting higher LGB identification than older groups.
You can review methods and the latest estimates in the ONS bulletin on sexual orientation, UK: 2023.
What Canadian Data Adds
Statistics Canada reports on sexual orientation through national surveys such as the Canadian Community Health Survey. Their releases often include breakdowns by sex and age, which helps with the “gay men vs lesbian women” question since you can compare within one sampling frame.
A solid entry point is the Statistics Canada release describing tables built from combined CCHS cycles: Socioeconomic profile of the 2SLGBTQ+ population (CCHS 2019–2021).
Why Different Sources Can Disagree
When one report says “more gay men” and another says “more lesbian women,” the clash often comes from design choices:
- Different places: the split can vary by country or region.
- Different age mix: a younger sample can tilt totals.
- Different label menus: adding labels like “queer” can spread responses across more options.
- Different mode: in-person, phone, and online collection can change skip rates.
- Different grouping rules: some analysts combine gay/lesbian with bisexual and some keep them separate.
None of that means the numbers are “made up.” It means the measurement details decide the headline.
Table: What Changes The Gay Vs Lesbian Count In Practice
| What Shifts The Number | How It Pushes Results | What To Do When Reading |
|---|---|---|
| Identity vs attraction vs behavior | Different measures can yield different totals | Match the headline to the measure used |
| Age band differences | Younger samples often report higher non-straight identity | Check age range and weighting notes |
| Label menu design | More options can spread responses across more labels | Read the response options, not just the chart |
| Mode of survey | Collection mode can change skip rates | Look for response-rate and missing-data notes |
| Small sample sizes | Wide uncertainty can flip close comparisons | Prefer pooled multi-year estimates when offered |
| Geography limits | Non-national samples can tilt totals | Use official household surveys for national shares |
So, Is There A Single Global Answer?
No single number works everywhere. The safest statement across countries is this: the gap between gay men and lesbian women is often small in identity surveys, and the ordering can change with sample design, age mix, and wording.
If you want the cleanest “apples to apples” read, stick to one country, one year, and one instrument. Compare men who selected “gay” with women who selected “lesbian” (or the combined “gay or lesbian” option reported by sex), then read the uncertainty notes.
Ways To Phrase The Takeaway Clearly
- “In this survey, gay men and lesbian women show up in similar ranges.”
- “This data set measures identity labels, not attraction or behavior.”
- “The estimate is close enough that sampling noise could switch the order.”
- “This result is for adults in this country in this year.”
If you want a long-running U.S. trend series in one place, see Gallup’s LGBTQ+ topic hub.
References & Sources
- CDC (NCHS).“NHIS – Sexual Orientation Information Frequently Asked Questions.”Explains how NHIS collects and reports sexual identity measures.
- Office for National Statistics (ONS).“Sexual orientation, UK: 2023.”Provides UK estimates by sex and age using the Annual Population Survey.
- Statistics Canada.“Socioeconomic profile of the 2SLGBTQ+ population (CCHS 2019–2021).”Describes tables built from combined CCHS cycles that include sexual orientation breakdowns.
- Gallup.“LGBTQ+ | Gallup Topic.”Tracks U.S. self-identification trends across years using national polling.